Each year, the U. S. National Institutes of Health (NIH) alone spends over $30 billion on medical research. The people who decide which scientists get this money are, unsurprisingly, scientists. Daniel Sarewitz, a professor of science and society at Arizona State University, thinks that we are no longer getting the bangs we should get for all these bucks. In an article in the spring/summer 2016 issue of The New Atlantis, he explains why.
First off, he admits that spending lots of money on scientific research and development has historically been a great idea. If you compare how most people in the U. S. lived in 1900 with the way things are done in 2000, most of the good differences—modern medicine, jet aircraft, cars, air conditioning, the Internet, wireless devices, computers, down to electric toothbrushes—are due to technological innovations that grew out of research directed at certain goals. And Sarewitz has no problem with that. The federal government was not a big player in research prior to World War II, but the lessons we learned then about how heavy investments in military technologies such as radar and nuclear weapons could pay off led to the creation of the National Science Foundation (NSF).
The NSF was the brainchild of MIT engineer Vannevar Bush, who directed most government-funded research during the war. Sarewitz says that in order to get his idea enacted, Bush told "a beautiful lie," and summarizes the lie this way: "Scientific progress on a broad front results from the free play of free intellects, working on subjects of their own choice, in the manner dictated by their curiosity for exploration of the unknown." Sarewitz spends the rest of his article showing just what was wrong about this "lie," and how it has led to inefficient and often simply incorrect research that taxpayers (and corporations too, for that matter) are paying billions for today.
In support of his thesis, he cites studies showing the increasing rate of retractions from peer-reviewed research journal in recent years. Another source indicates that between 75 and 90 percent of all basic and preclinical biomedical studies are not reproducible.
These are indicators of a general trend or pattern that goes something like this. A newly minted Ph. D. in one of the softer sciences (economics, sociology, psychology, biology) gets an academic job and has to produce new and original results that are published in peer-reviewed journals or get fired in five or six years. (That part I'm very familiar with.) So he writes tons of proposals to NIH or wherever he can get funding, gets ten to fifteen percent of his proposals funded, and sets to work being novel. Novel about what? Well, that almost doesn't matter. As long as you can show you did something that nobody's done before, and it has some remote tenuous connection with reality, you can find a journal and willing referees to publish it. There's more Internet-based journals popping up every year, and it's almost to the point that you can get anything published if you send it to enough journals. Multiply this picture by the thousands of Ph. D.'s we produce every year, and bear in mind that their proposals and papers are being reviewed by people who went through the very same system, and you can see how the situation described by Sarewitz can happen.
So what should we be doing with all that money? Sarewitz says it should be spent on highly directed research targeted at specific time-limited goals. He cites several examples of the way it ought to be done, including a program in the 1990s directed by the U. S. Department of Defense (DOD) to develop a new treatment for breast cancer. It led to the development of the drug herceptin, one of the most important treatment innovations in years. The point here is that useful innovations based on science typically happen not when some isolated scientist pursues his or her private dream, but when a team of smart people use both existing and new science and engineering to pursue a specific socially profitable goal.
Sarewitz congratulates the DOD for its no-nonsense approach to getting such things done, and for including in its project planning people without academic qualifications, but with a strong interest in the goal. One of the sparkplugs that got the herceptin project going was an activist named Fran Visco, herself a breast cancer survivor, who founded the National Breast Cancer Coalition to do something about better treatments. She was a lawyer, not a biologist, but the DOD welcomed her as a participant and her vision was essential in getting things done.
In pointing out that much research today, especially in the biomedical area, doesn't seem to accomplish much more than paying for a lot of expensive professors, postdocs, grad students, and equipment, I think Sarewitz is on the money, so to speak. However, I disagree with him that Vannevar Bush told "a beautiful lie" to get the NSF going.
If you genuinely believe that what you say at the time is true, it isn't a lie, morally speaking. You may be guilty of self-deception or not informing yourself sufficiently, but not of lying. Bush was a creature of his time, and most of the people he had hired to develop things like radar and nuclear bombs had in fact spent most of their careers indulging the "free play of their free intellects," because that was the main way basic science was pursued before huge amounts of federal dollars showed up after World War II. Most scientists before 1941 operated more or less in the style of Albert Einstein, who single-handedly revolutionized physics in his spare time that his job at the Swiss patent office provided him. Back then, the exception was a scientist who did good science while working for industry, such as Irving Langmuir, the first industrial scientist to win a Nobel Prize. So although the phrase "beautiful lie" is an attention-getting rhetorical device, I think Sarewitz is a little anachronistic in his accusation that Bush was lying at the time.
On the other hand, Sarewitz is probably right in pointing out that the foxes-guarding-the-henhouse pattern of handing money over to scientists who give it to other scientists may not be the right way to do things anymore, at least with the majority of federal research funds. There's a reason that government funding for science is declining as a percent of GNP, and the public may be right in thinking that their federal research dollars are not being spent as wisely as they could be. If the powers that are or will be listen to Sarewitz's advice, maybe things will be reorganized so that even fewer dollars can accomplish more, both in the way of pure basic science and in practical applications that improve the lives of millions.
